How Accurate Are Coaches in Predicting Perceived Exertion of Their Athletes?

Abstract Ryan, GA, Bunn, JA, and Eisenman, ML. How accurate are coaches in predicting perceived exertion of their athletes? J Strength Cond Res 37(3): 684–687, 2023—Ratings of perceived exertion (RPE) are used extensively to subjectively evaluate workload in sport. The purpose of this study was to compare season long player to coach perceptions of RPE loads (RPEL) in Division I men's soccer and women's lacrosse programs. Coach and player self-reported RPE were collected during the competitive seasons for both teams within 30 minutes of training sessions. Scores of RPE were multiplied by recorded training time to determine RPEL and averaged to create a daily coach and player RPEL. Accuracy of RPEL prediction between coaches and players was determined via mean bias, mean absolute percent error (MAPE), correlation, and an equivalence test. Soccer coaches underestimated RPEL (1,345 ± 410 arbitrary unit [AU]) compared with players (1,372 ± 435 AU). The coach mean bias was −27.3 ± 151.9 AU, wieth an MAPE of 7.4%, a strong correlation ( r = 0.937, p < 0.001), and coach to player equivalence ( p = 0.045). Lacrosse coaches overestimated RPELs (876 ± 349 AU) compared with players (829 ± 214 AU). The coach mean bias was 47 ± 189 AU, with an MAPE of 14.7%, a strong correlation ( r = 0.883, p < 0.001), and coach to player equivalence ( p = 0.010). Analyses indicated that coaches accurately predicted player RPEL within both sports. Variance occurred on days with high RPELs, with overestimations occurring in lacrosse and underestimation in soccer. Coaches can use RPELs to assess subjective training load but should consider objective measures, including microtechnology, to provide a comprehensive assessment.